3 ms·

I think the main difference between the two kinds of forks is that in one case existing holders of Bitcoin get to own both the original and the fork, and in the

I think the main difference between the two kinds of forks is that in one case existing holders of Bitcoin get to own both the original and the fork, and in the other case the fork must bootstrap itself from scratch without the added benefit of a bunch of biased people who are happy to see any new speculative endeavor grow (as long as they own some). Hence, hard forks are actually much more likely to occur, at least until the inverse network effects you describe actually exist in reality. In a highly speculative market they really don't exist, or at least don't exist enough to discourage a fair bit of aggressive forking. I predict at least two hard forks in 2018.

philipodonnell 9y ago
> hard forks are actually much more likely to occur, at least until the inverse network effects you describe actually exist in reality. I am working on an article that talks about the opposing incentives behind hard forks as a form of artificial scarcity. This seems to be the main objection from classical financial types, that its all virtual and you can make more at any time by forking so its worthless. Your comment implies thoughtfulness on this topic. Do you have any good references on this from an economic theory standpoint? I could argue that there are actually two kinds of hard forks, one where there is some change to the underlying protocol (Bitcoin Cash) or reallocation of funds (undoing the DAO hack) that, at least in the eyes of the authors, increases the utility of the new network more than the negative effect of the split. If the total network value rises then, in a rational market, it means that the utility did increase. A second kind of fork is just a cash-grab fork. Split the bitcoin chain but don't do anything to the protocol, ala https://forkgen.tech https://forkgen.tech. This is (I think) what concerns finance professionals more than forks which increase utility.
- olfactory 9y ago
  > I could argue that there are actually two kinds of hard forks, one where there is some change to the underlying protocol (Bitcoin Cash) or reallocation of funds (undoing the DAO hack) that, at least in the eyes of the authors, increases the utility of the new network more than the negative effect of the split. If the total network value rises then, in a rational market, it means that the utility did increase. I think this is exactly correct from an economic perspective. But I think that in the heavily speculative market it will take a while for the market to actually converge into the state that you describe. I'd argue that right now there is virtually no disincentive for doing hard forks, except perhaps for the friction entailed in getting it listed on Coinbase. If a hard fork "launches" with a low price, many speculators will simply view it as a buy opportunity, whereas if it launches with a high price, it has already achieved some level of legitimacy. This is because hard forks that are effectively governance changes are not really purchased for fundamentals-based reasons at this point, because the governance aspects of BTC have really not been tested much to date. I actually have in mind a few very interesting hard forks if you would like to collaborate and attempt to make some money by launching these with me in the process :) I too am interested in it from a research perspective but it's very tempting to try to "add value" by launching some important hard forks that the community hasn't thought of yet.
  - thisisit 9y ago
    IMO, what happened with Ethereum was not exactly the correct economic perspective. More like the ethereum devs went on to support the Ethereum and ditched the coin. Bitcoin Cash/Gold on the other hand seems fine. > except perhaps for the friction entailed in getting it listed on Coinbase And therein lies the rub. I'd say its not only Coinbase but exchanges in general. My understanding is that even the worse of exchanges like HitBtc charge 25 BTC to list a coin. So, if you have plans to make at least 25-30 BTC to get through the exchange doors, count me in :P
    - olfactory 9y ago
      > IMO, what happened with Ethereum was not exactly the correct economic perspective. More like the ethereum devs went on to support the Ethereum and ditched the coin. Bitcoin Cash/Gold on the other hand seems fine. I agree with this, actually. The hard fork showed that ETH would be forcibly redistributed if the result of correct VM behavior was not what the maintainers had in mind. So the ETH hard fork fits into the very typical pattern in human institutions of excessive centralization resulting in corrupt behavior. Most of the early adopters of ETH who lost money due to the DAO attack were essentially the political allies of the maintainers, who in turn held great sway with miners. We've recently learned that ETH mining is far more centralized than BTC. The ETH hard fork was a nice wakeup call that even for supposedly enlightened proponents of decentralization and distributed consensus, when there is real money on the line pretty much any excuse will be acceptable for why the "theft" had to be unwound via a hard fork. Ironically, in order for market incentives to work properly in a smart contract system, finding exploitable cases of correct (or incorrect) behavior of the VM should result in profit, or else there is simply no market incentive for the system to be secure and to behave in a predictable manner. In order for Ethereum to have passed the institutional test imposed on it by the DAO hack, it would have had to deal with the difficult issue of victims of the attack claiming "theft" and would have had to allow the difficult lesson to be learned. Now that we know that Ethereum was a toy when the DAO hack occurred, when will it stop being a toy? Is the current fork real? Anyway, apologies for the hyperbole, I just wanted to make the above point.
  - philipodonnell 9y ago

philipodonnell 9y ago

> hard forks are actually much more likely to occur, at least until the inverse network effects you describe actually exist in reality. I am working on an article that talks about the opposing incentives behind hard forks as a form of artificial scarcity. This seems to be the main objection from classical financial types, that its all virtual and you can make more at any time by forking so its worthless. Your comment implies thoughtfulness on this topic. Do you have any good references on this from an economic theory standpoint? I could argue that there are actually two kinds of hard forks, one where there is some change to the underlying protocol (Bitcoin Cash) or reallocation of funds (undoing the DAO hack) that, at least in the eyes of the authors, increases the utility of the new network more than the negative effect of the split. If the total network value rises then, in a rational market, it means that the utility did increase. A second kind of fork is just a cash-grab fork. Split the bitcoin chain but don't do anything to the protocol, ala https://forkgen.tech https://forkgen.tech. This is (I think) what concerns finance professionals more than forks which increase utility.

olfactory 9y ago
> I could argue that there are actually two kinds of hard forks, one where there is some change to the underlying protocol (Bitcoin Cash) or reallocation of funds (undoing the DAO hack) that, at least in the eyes of the authors, increases the utility of the new network more than the negative effect of the split. If the total network value rises then, in a rational market, it means that the utility did increase. I think this is exactly correct from an economic perspective. But I think that in the heavily speculative market it will take a while for the market to actually converge into the state that you describe. I'd argue that right now there is virtually no disincentive for doing hard forks, except perhaps for the friction entailed in getting it listed on Coinbase. If a hard fork "launches" with a low price, many speculators will simply view it as a buy opportunity, whereas if it launches with a high price, it has already achieved some level of legitimacy. This is because hard forks that are effectively governance changes are not really purchased for fundamentals-based reasons at this point, because the governance aspects of BTC have really not been tested much to date. I actually have in mind a few very interesting hard forks if you would like to collaborate and attempt to make some money by launching these with me in the process :) I too am interested in it from a research perspective but it's very tempting to try to "add value" by launching some important hard forks that the community hasn't thought of yet.
- thisisit 9y ago
  IMO, what happened with Ethereum was not exactly the correct economic perspective. More like the ethereum devs went on to support the Ethereum and ditched the coin. Bitcoin Cash/Gold on the other hand seems fine. > except perhaps for the friction entailed in getting it listed on Coinbase And therein lies the rub. I'd say its not only Coinbase but exchanges in general. My understanding is that even the worse of exchanges like HitBtc charge 25 BTC to list a coin. So, if you have plans to make at least 25-30 BTC to get through the exchange doors, count me in :P
  - olfactory 9y ago
    > IMO, what happened with Ethereum was not exactly the correct economic perspective. More like the ethereum devs went on to support the Ethereum and ditched the coin. Bitcoin Cash/Gold on the other hand seems fine. I agree with this, actually. The hard fork showed that ETH would be forcibly redistributed if the result of correct VM behavior was not what the maintainers had in mind. So the ETH hard fork fits into the very typical pattern in human institutions of excessive centralization resulting in corrupt behavior. Most of the early adopters of ETH who lost money due to the DAO attack were essentially the political allies of the maintainers, who in turn held great sway with miners. We've recently learned that ETH mining is far more centralized than BTC. The ETH hard fork was a nice wakeup call that even for supposedly enlightened proponents of decentralization and distributed consensus, when there is real money on the line pretty much any excuse will be acceptable for why the "theft" had to be unwound via a hard fork. Ironically, in order for market incentives to work properly in a smart contract system, finding exploitable cases of correct (or incorrect) behavior of the VM should result in profit, or else there is simply no market incentive for the system to be secure and to behave in a predictable manner. In order for Ethereum to have passed the institutional test imposed on it by the DAO hack, it would have had to deal with the difficult issue of victims of the attack claiming "theft" and would have had to allow the difficult lesson to be learned. Now that we know that Ethereum was a toy when the DAO hack occurred, when will it stop being a toy? Is the current fork real? Anyway, apologies for the hyperbole, I just wanted to make the above point.
- philipodonnell 9y ago